PSMA is not a new protein and we’ve known of its existence for a long time. However, we were not able to use it for treatment or even imaging until quite recently.

PSMA is a transmembrane protein that is expressed by prostate cancer cells and associated metastases¹; in metastatic prostate cancer, it is overexpressed in 90% of cases.² PSMA is rarely expressed by normal tissues, with the exception of the salivary glands and the kidneys. The expression of PSMA in prostate cancer correlates with disease burden and aggressiveness, so we see more PSMA expression in mCRPC. Usually, there is a lower expression in liver metastases.²

As mentioned previously, the PSMA protein is clinically important for both imaging purposes and treatment strategies. PSMA‑PET has been available for more than a decade. It is clearly more sensitive than other imaging modalities such as bone scans, CT scans, and other PET scans that we’re using in men with prostate cancer.³ Like any other imaging, PSMA-PET can have false positives and false negatives. It is still important to critically evaluate the imaging and ensure that what you are seeing is actually true; of course, this is challenging.

Having said that, it’s very clear that we see many more lesions on PSMA‑PET compared with what a bone scan would report, for example. Within the past decade, we have shown clearly that 2 drugs^4-8 and, likely now, 3 drugs are needed to treat men with metastatic castration-sensitive prostate cancer (mCSPC) with the combination of ADT, docetaxel chemotherapy, and a next-generation hormonal agent (darolutamide or abiraterone).^9,10 However, we don’t know whether these data still apply when metastases are evident by PSMA‑PET and not a bone scan or a CT scan. There is an ongoing debate regarding those issues.

Lutetium 177 vipivotide tetraxetan PSMA targeted radioligand therapy (¹⁷⁷Lu-PSMA-617 is a new treatment for prostate cancer. In this slide, the transmembrane protein is depicted in green and the lutetium PSMA radioligand is depicted in purple. Lutetium 177 is linked to the ligand of PSMA and is what is injected into the patient’s blood. ¹⁷⁷Lu-PSMA-617 binds to the transmembrane protein at the cell surface and undergoes endocytosis. The radioligand/protein couple enters the cell where lutetium 177 delivers radiation to the cell and induces DNA damage, which leads to cell death.¹¹

¹⁷⁷Lu-PSMA-617 was initially used in Germany nearly 10 years ago, where it was tested on a compassionate use basis. Some successes were reported by physicians as well as noted on imaging, while patients reported an improvement in pain.¹² Our Australian colleagues were able to conduct a nonrandomized, prospective, single-arm phase II trial that showed clear evidence of efficacy with ¹⁷⁷Lu-PSMA-617.¹³ They then followed their program with a randomized, open-label phase II trial of ¹⁷⁷Lu-PSMA-617 compared with cabazitaxel.¹⁴ From there, we were able to conduct an international, open-label, randomized phase III trial of ¹⁷⁷Lu-PSMA-617 in combination with standard of care (SoC) compared with SoC alone, which resulted in the FDA approval of ¹⁷⁷Lu-PSMA-617 for the treatment of PSMA-positive mCRPC previously treated with androgen receptor pathway inhibition and taxane-based chemotherapy.^11,15

The investigators of this study selected their patients using a stringent strategy. Investigators used PSMA‑PET for selection, and patients had to have evidence of strong PSMA expression in their cancers. They also used fluorodeoxyglucose (FDG)‑PET to ensure lesions were not PSMA negative. Patients enrolled in this study were in the best situation for the treatment to work. The prostate-specific antigen (PSA) response waterfall plot shows a very high response rate being reported with 57% of patients experiencing a PSA decline of ≥50% (PSA50).¹³

Investigators then decided to initiate a randomized, open-label phase II trial comparing ¹⁷⁷Lu-PSMA-617 with cabazitaxel (TheraP trial). Men with mCRPC who had failed existing therapies such as docetaxel, abiraterone, and enzalutamide were included. Similar to the previous phase II trial, patients were selected based on both PSMA‑PET and FDG‑PET, again indirectly favoring ¹⁷⁷Lu-PSMA-617 because the men who were selected were most likely to benefit from ¹⁷⁷Lu-PSMA-617.

A total of 37% of men who received cabazitaxel achieved PSA50, which is an expected response for cabazitaxel in this setting. In the ¹⁷⁷Lu-PSMA-617 arm, 66% of patients experienced a PSA50 response (95% CI: 16-42; P <.0001).¹⁴

TheraP also looked at progression‑free survival (PFS) and overall survival (OS). However, because this was a randomized phase II trial and not a phase III trial, it was not designed to show superiority of one arm over the other. Cabazitaxel is more commonly being used as a benchmark for efficacy. Results showed that ¹⁷⁷Lu-PSMA-617 appeared slightly better than cabazitaxel in terms of PFS. In terms of OS, the 2 arms were incredibly similar and no difference in OS was noted with an HR of 0.97 (95% CI: 0.70-1.40; P = .99).¹⁶

I think what we’re seeing is a clear gain for ¹⁷⁷Lu-PSMA-617 efficacy in a very select population that is most likely to benefit from the therapy. Cabazitaxel still works in the same population, which is very good to know. The question is not necessarily whether we should use ¹⁷⁷Lu-PSMA-617 instead of cabazitaxel, but rather, how best to select patients for one of those 2 treatments. Perhaps of even more importance, should we combine these 2 treatments? Should we sequence them? Can we find biomarkers that predict for efficacy for one vs the other? In my opinion, those are the most important questions rather than trying to assess whether one drug is better than the other.

The VISION study is the first phase III trial testing ¹⁷⁷Lu-PSMA-617 in men with mCRPC. Similar to the previous studies, VISION selected men with very advanced disease. Patients had to have received at least 1 taxane and at least 1 androgen receptor pathway inhibitor to be included. Patients were required to have a performance status of 0, 1, or 2 as well as adequate major organ and bone marrow function. Of most importance, PSMA‑PET was requested so patients with PSMA‑positive cancer would be selected for the trial.

Patients were randomized in a 2:1 fashion to receive protocol‑permitted SoC alone or in combination with ¹⁷⁷Lu-PSMA-617 once every 6 weeks for 4 cycles. ¹⁷⁷Lu-PSMA-617 could be extended up to 6 cycles in men who were benefiting. Best SoC included either a second androgen receptor pathway inhibitor, radiation therapy, dexamethasone, or supportive care. Best SoC encompassed essentially everything you could give this population of men with advanced disease, with the exception of chemotherapy or radium-223. We did not have data supporting the combination of ¹⁷⁷Lu-PSMA-617 plus chemotherapy or radium-223. 

The 2 primary endpoints were radiographic PFS and OS. Key secondary endpoints included time to first symptomatic skeletal event and quality of life (QoL).¹⁵

Nearly 2000 patients were screened for the VISION study. Approximately 15% of these patients did not have access to PSMA‑PET or experienced cancer progression before they could have access to the imaging. PSMA criteria were met in the majority of those who received a PSMA-PET (86.6%), whereas 12.6% of patients had no expression of the PSMA protein (PSMA-PET negative) or there was a discrepancy between the PSMA‑PET and the CT scan. A discrepancy between PSMA-PET and CT was noted as lesions seen on CT that would not express the PSMA protein when assessed on PSMA-PET.

A total of 831 patients were randomized to SoC with or without ¹⁷⁷Lu-PSMA-617.¹⁵

All patients had to have received an androgen receptor pathway inhibitor; of note, this study was conducted before darolutamide was available. Almost all patients had received docetaxel (96.9%) and approximately 40% of them had also previously received cabazitaxel.

Nearly 46% of patients had received >1 androgen receptor pathway inhibitor and 41% of the population had received 2 taxanes: docetaxel and cabazitaxel. Again, this was a population with very advanced cancer.¹⁵

What was suggested by the nonrandomized phase II trial conducted in Australia was able to be confirmed in VISION. A total of 46% of men who received best SoC plus ¹⁷⁷Lu-PSMA-617 achieved a ≥50% decline in PSA as opposed to only 7.1% of those who received best SoC alone.¹⁵

The results of the trial showed that ¹⁷⁷Lu-PSMA-617 was clearly superior, with 8.7 months being the median radiographic PFS for ¹⁷⁷Lu-PSMA-617 as opposed to 3.4 months in the control arm without ¹⁷⁷Lu-PSMA-617 (HR: 0.40; 99.2% CI: 0.29-0.57; P <.001). It is evident by the shape of the PFS curves that this benefit is present and appears quite fast after approximately 3 months, and it is maintained until approximately 1.5 years before almost all patients eventually progress. This disease, unfortunately, remains noncurative for even the best population of men, but this is a tremendous benefit with a 60% reduction in the risk of progression or death seen with the addition of ¹⁷⁷Lu-PSMA-617.¹⁵

¹⁷⁷Lu-PSMA-617 had a clear impact on OS with a median OS of 15.3 months in the ¹⁷⁷Lu-PSMA-617 arm vs 11.3 months in the control arm (HR: 0.62; 95% CI: 0.52-0.74; P <.001), which led to a 38% reduction in the risk of death. It does not take long to start to see the benefit of ¹⁷⁷Lu-PSMA-617, with the curves beginning to separate at 2-3 months and remaining parallel until more than 2 years of follow-up.¹⁵

Results of a subgroup analysis were recently released to determine the OS benefit of ¹⁷⁷Lu-PSMA-617 with previous therapies received. To be enrolled in the trial, patients were required to have received at least 1 androgen receptor pathway inhibitor and 1 taxane, but patients may have received several other treatments including bone health agents or radium-223. The OS benefit of ¹⁷⁷Lu-PSMA-617 when added to best SoC was consistent across all previous treatment subgroups.¹⁷

Another subgroup analysis was performed to determine the benefit of ¹⁷⁷Lu-PSMA-617 combined with different concomitant SoC treatments used in the study. As mentioned previously, the SoC was quite inclusive in this trial to reflect real-world practice. Consistent OS benefits were seen with the ¹⁷⁷Lu-PSMA-617 plus SoC arm, regardless of which concomitant therapies were used as part of the SoC.¹⁷

At the 2021 European Society for Medical Oncology (ESMO) Congress, QoL and pain assessments from VISION were reviewed. Patients who received ¹⁷⁷Lu-PSMA-617 were noted to have a statistically significant longer time to worsening in QoL compared with SoC alone, with an HR of 0.46 (95% CI: 0.35-0.61; P <.001). The median time to worsening was approximately 9.7 months with ¹⁷⁷Lu-PSMA-617 as opposed to only 2.4 months in the control arm, which means you’re giving prolonged quality time to your patients with this therapy. ¹⁷⁷Lu-PSMA-617 not only maintains QoL but it also prevents the worsening of QoL compared with what we would normally do for these men.

One caveat to be aware of is that QoL data were not available for many of the patients due to their lack of response in completing the questionnaires. Of course, this is a very frequent limitation when QoL is assessed in studies.

Many of these men will have bone pain due to their bone metastases and the Brief Pain Inventory–Short Form (BPI‑SF) questionnaire was used in this trial to assess pain. The HR for the BPI‑SF pain intensity was 0.45 (95% CI: 0.33-0.60; P <.001) favoring ¹⁷⁷Lu-PSMA-617, with a median time to worsening of >1 year (14.3 months) vs only 2.9 months in the control arm. This is a very big difference indicating what this agent can provide to patients in terms of symptom control and QoL.¹⁸

Another very important endpoint addressed in VISION is time to symptomatic skeletal events. I’m speaking of fractures, major pain requiring radiation, lesions at risk requiring surgery, or spinal cord compression—significant events that we’re fearing for our patients. Here again, the benefit of ¹⁷⁷Lu-PSMA-617 was tremendous, in my opinion. The HR for time to first symptomatic skeletal event was 0.5 (95% CI: 0.40-0.62; P <.001), which is very rarely seen when it comes to symptomatic skeletal events; the median was 11.5 months in the ¹⁷⁷Lu-PSMA-617 arm vs 6.8 months in the control arm. This outcome is 2 times better with ¹⁷⁷Lu-PSMA-617 in terms of preventing severe bone-related events from the disease, which is remarkable.¹⁵

What is the price to pay for the patients to get this benefit? There was an excess in fatigue in the ¹⁷⁷Lu-PSMA-617 arm (49.1% vs 29.3%). It is important to remember that patients in the ¹⁷⁷Lu-PSMA-617 arm were followed longer because they lived longer. Due to that, the likelihood to report fatigue is higher, so it doesn’t necessarily mean that the treatment is directly associated with fatigue.

Another very important toxicity of note is bone marrow suppression. A total of 23.4% of patients experienced grade 3-5 bone marrow suppression with ¹⁷⁷Lu-PSMA-617 vs 6.8% with SoC alone. 1 patient experienced grade 5 bone marrow failure in the ¹⁷⁷Lu-PSMA-617 group. This is important to remember and is likely due to irradiation of the bone marrow by the radioisotope.

VISION included patients with very advanced disease. It was common for patients to have numerous bone metastases. The amount of bone disease was enormous in the trial. It leads to a bit of a debate as to whether a man with a very large bone disease burden is a good candidate for ¹⁷⁷Lu-PSMA-617. On the one hand, these patients may benefit tremendously because men with bone metastases are at risk of severe bone morbidity. On the other hand, they may develop bone marrow suppression, which may require transfusion support to mitigate the bone marrow suppression. I don’t have a final answer or recommendation to give regarding this debate, but I think all of us should be aware of it and do our best to make the decision on a patient-by-patient basis.

The third significant side effect associated with ¹⁷⁷Lu-PSMA-617 is dry mouth, which was seen in approximately 40% of men receiving ¹⁷⁷Lu-PSMA-617 vs only 1% in the control arm. This was a well-known side effect of ¹⁷⁷Lu-PSMA-617, which can be explained by the mechanism of action. The PSMA protein is expressed by salivary glands; as a result, we are irradiating the salivary gland, which can lead to dry mouth. The dry mouth associated with ¹⁷⁷Lu-PSMA-617 is usually mild. In my experience with my own patients, it does not affect their ability to eat, feel, or taste and it does not get worse with repeated injections. Despite the high incidence of dry mouth, it does not appear to have a big clinical impact for patients.

Finally, I want to mention that this treatment is associated with nausea and vomiting in 39.3% of patients vs 17.1% in the control arm.¹⁵ This is a preventable side effect. These patients should receive ondansetron and corticosteroids, which is also my standard practice.

This is a recent paper reported by investigators at the 2022 Annual Meeting of the American Society of Clinical Oncology (ASCO), who sought to determine the outcomes of patients treated with ¹⁷⁷Lu-PSMA-617 who did not meet VISION criteria for imaging. In this retrospective analysis, the PFS and OS data for patients who met VISION’s imaging criteria were similar to that shown in the VISION trial. For those who did not meet VISION’s imaging criteria, the outcome for these men was quite poor with a median PFS of 2.1 months and 9.6 months for the median OS.¹⁹

The question we do not have an answer to is whether this is related to biology. In other words, we do not know whether men with PSMA‑negative cancers simply have a worse cancer or whether this is related to a discrepancy between PETs. For example, if a man has liver metastases that don’t express PSMA, he may experience a bad outcome simply because he has liver metastases, not necessarily because PSMA is not expressed. We don’t have the answer to this question and, hopefully, there will be more analyses and studies coming to enlighten us regarding this question.

¹⁷⁷Lu-PSMA-617 is not the only means of targeting PSMA. For example, PSMA bispecific antibodies have been developed in the past 2‑3 years and there is promise with this treatment.²

Bispecific antibodies are essentially smart immunotherapy. Think of these treatments as an antibody with 2 arms. 1 arm is targeting CD3 on the T-cells while the other arm will bind to PSMA on tumor cells. This leads to T-cell activation and tumor cell death.²⁰

AMG 160 is a bispecific antibody targeting PSMA. AMG 160 has an extended half-life that allows for longer dosing intervals between administrations.² Results of an open-label phase I study were presented at ESMO 2020. It’s clearly an active agent with a PSA reduction seen in the majority of men with mCRPC who have already exhausted at least 1 androgen receptor inhibitor and 1 taxane. As indicated by the waterfall plot, 34.3% of men achieved a PSA50 response.²¹

Similar to ¹⁷⁷Lu-PSMA-617, dry mouth is a common side effect with approximately one third of men who received AMG 160 experiencing this side effect. The main concern with AMG 160 is the incidence of CRS, which was seen in 90.7% of patients with one fourth of patients developing grade ≥3 CRS. CRS is associated with fever, low blood pressure, and significant fatigue and can require ICU-level care for monitoring, hydration, and blood pressure support. This makes bispecific T-cell engager treatments more challenging to be used in the outpatient setting.²¹

There have been attempts to circumvent CRS, and I think more data will be presented soon regarding that. Potential techniques include pretreatment hydration and prophylaxis early management as well as drug administration with dose titration or continuous infusion. 

HPN424 is another PSMA-directed T-cell engager. HPN424 has 3 binding domains: CD3 (T-cell engagement), PSMA (tumor engagement), and albumin (half-life extension). Preliminary data from a phase I/IIa study were reported at ASCO 2021 with HPN424 showing clear efficacy. The maximum tolerated dose has not yet been reached and there’s a suggestion that higher doses might be associated with greater efficacy.²² I am looking forward to more data with this agent to better assess its efficacy and, also, its toxicity.

There are many other drugs in development that aim to target PSMA. Some others being investigated are radionuclides like ¹⁷⁷Lu-PSMA-617, which I discussed above. For example, actinium PSMA is an α-emitting radionuclide currently under development. α emitters may have a higher potential for efficacy, but potentially also higher toxicity to the salivary glands, so we will need to figure out how best we should develop α emitters targeting PSMA in this regard.²

I mentioned bispecific and trispecific antibodies in the form of AMG 160 and HPN424. AMG 160 is currently being investigated in earlier lines of therapy. A phase I/II study is looking at AMG 160 as monotherapy or in combination with novel hormonal therapies (enzalutamide or abiraterone) or PD-1 inhibition (AMG 404) in patients with mCRPC before they receive taxane-based chemotherapy. REGN5678 is a bispecific antibody targeting PSMA and CD28. It is currently being investigated in a phase I/II trial in combination with PD-1 inhibitor cemiplimab in patients with mCRPC. We should see more data in future congresses regarding these studies (NCT03792841, NCT04631601, NCT03972657).

CAR T-cell therapies have also been tested in this setting and we have preliminary reports indicating efficacy. 5 patients received anti-PSMA CAR T-cells in combination with interleukin-2 after chemotherapy conditioning. 2 patients achieved a partial response with PSA decreases of 50% and 70%.²³ In vivo models have shown increased proliferation of T-cells, enhanced tumor eradication, and improved cytokine secretion with coexpression of TGFβRdn on PSMA-directed CAR T-cells in metastatic prostate cancer. A phase I study of CART-PSMA-TGFβRdn is underway and results are anxiously awaited.²⁴ Of course, the use of CAR T-cells comes with great complexity, as we’ve seen in other disease states.²

Antibody–drug conjugates (ADCs) combine a monoclonal antibody with a cytotoxic agent. We have been waiting for ADCs in prostate cancer for quite some time; I am glad to see that there are PSMA ADCs under development that have seen promising activity. MEDI3726 is a PSMA-targeted ADC that was tested in a phase I study of men with mCRPC after progression on enzalutamide and/or abiraterone as well as taxane-based chemotherapy. The median PFS and OS were 3.9 months and 10.6 months, respectively. However, 90.9% of patients experienced drug-related adverse events, with 39.4% of patients discontinuing therapy as a result.²⁵

There are different philosophies when it comes to imaging selection. One philosophy is to select patients not only with evidence of PSMA protein in their cancer, but also to exclude patients with any lesions that do not express PSMA. This is the most extreme way to select and is done by performing a PSMA‑PET as well as an FDG‑PET and correlating the 2 imaging modalities before decision making.^13,14

Another philosophy questions if it’s necessary to scan patients before decision making given that the protein is expressed in the majority of cases.

In the middle of those 2 ends of the spectrum is a third philosophy that was utilized in VISION. Patients had a PSMA‑PET scan, but also, the second imaging was done using simply a CT scan, which is cheaper and easier to access than other PET imaging scans generally are.¹⁵

I don’t have a final answer as to which imaging strategy is ideal to select, but just be aware there are different possibilities.

As always in oncology, the earlier use of therapeutic agents is tempting. There are several planned and ongoing trials testing PSMA targeting; I think the most important phase III trials are trying to establish the role of ¹⁷⁷Lu-PSMA-617 in an earlier setting.

The PSMAddition study is investigating ¹⁷⁷Lu-PSMA-617 in men with mCSPC for the first time and I think this is very important. PSMA is expressed quite early in the course of the disease. It is unknown whether we can improve patient outcomes in this setting with PSMA targeting (NCT04720157). Our current standards use 2 or even 3 drug regimens for these men; we don’t know at this point whether PSMA targeting should be part of that.

For men with mCRPC who have not received taxane‑based chemotherapy, 2 phase III trials called PSMAfore and SPLASH are asking whether lutetium 177 should be used earlier in therapy (NCT04689828, NCT04647526).

Finally, a phase III trial called PROSTACT is looking to confirm the data from VISION in the post‑docetaxel mCRPC setting using ¹⁷⁷Lu-DOTA-rosopatamab (NCT04876651).

Regarding conversations with patients considering ¹⁷⁷Lu-PSMA-617, I start with very practically explaining how the imaging works, how the treatment works, and associated adverse events. If the patient is a candidate for treatment with ¹⁷⁷Lu-PSMA-617 using the VISION criteria, I move on to discussing treatment location and length because this is administered intravenously every 6 weeks in the nuclear medicine department.

If the patient doesn’t meet VISION criteria for ¹⁷⁷Lu-PSMA-617, I explain how active the treatment is and that we believe it may have its place earlier in the course of the disease. To use the phase III PSMAfore trial as an example, if a patient has exhausted an androgen receptor drug, that patient could be considered for chemotherapy treatment. However, if the patient is not willing to receive chemotherapy at the time, the patient may be eligible for a phase III trial randomizing ¹⁷⁷Lu-PSMA-617 either now or later on if the cancer progresses while on a second androgen receptor or hormonal therapy.

If a patient has mCSPC, I explain the current SoC and if the cancer expresses the PSMA protein, I may consider using ¹⁷⁷Lu-PSMA-617 as part of a randomized clinical trial. 

Enrollment in a clinical trial for AMG 160 has potential to be an option for patients who have already received ¹⁷⁷Lu-PSMA-617. Thinking about the investigational bispecific antibodies, there were select patients in the phase I/II trial of AMG 160 who benefited on an individual basis from AMG 160 after previously receiving ¹⁷⁷Lu-PSMA-617. So, at least preliminarily, it appears that this strategy is doable and some individuals may benefit. We need more data to be certain of a broader population benefit.